AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

DeCap: Decoding CLIP Latents for Zero-Shot Captioning via Text-Only Training

Wei Li, Linchao Zhu, Longyin Wen, Yi Yang

ICLR 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 24 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Project the image embedding into the text embedding space instead of using the visual embedding directly (ablation)DeCap · MSCOCO, in-domain unpairedCIDEr: 25.7 -> 91.2Improved
Decode the nearest text embedding rather than retrieve its original caption (ablation)CLIPRe / DeCap-NND · MSCOCO, in-domain unpairedCIDEr: 53.4 -> 62.9Improved
Train the decoder on SS1M rather than CC3M-textDeCap · MSCOCO Karpathy-test, zero-shotCIDEr: 42.1 -> 50.6Improved
Add target-domain text to construct the support memory (ablation)DeCap · Flickr30K to MSCOCO, cross-domainCIDEr: 44.4 -> 63.1Improved
Train the decoder on the full MSCOCO text set instead of 1% of it (ablation)DeCap · MSCOCO Karpathy-testCIDEr: 81.5 -> 91.2Improved
Use a similarity-filtered support memory instead of randomly sampling the same number of embeddings (ablation)DeCap · MSCOCO Karpathy-testCIDEr: 38.2 (-4.0) -> 42.3 (+0.1)Improved
Use decoder-reconstructed captions as the support memory (ablation)DeCap · MSCOCO Karpathy-testCIDEr: 91.2 -> 95.1 (+3.9)Improved
Add the prompt “Attention! There is/are” (ablation)DeCap-BookCorpus · MSCOCO Karpathy splitCIDEr: 21.8 -> 31.9 (+10.1)Improved
Add the prompt “Attention! There is/are” (ablation)DeCap-CC3M · MSCOCO Karpathy splitCIDEr: 42.1 -> 42.2 (+0.1)No clear effect
Fine-tune on MSCOCO text after pre-training on CC3M text (ablation)DeCap · NoCaps val, overallCIDEr: 38.3 -> 58.2Improved
Switch from CC3M pre-training to MSCOCO pre-training (ablation)DeCap · NoCaps val, out-of-domainCIDEr: 48.7 -> 25.8Hurt
Replace random support sampling with a 1% support memory (ablation)DeCap · MSCOCO Karpathy-testCIDEr: 3.8% performance drop in CIDErHurt

Later papers that compared with this method

6 later accepted papers report DeCap as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy